Don't panic when tasting a disk with sectorsize=0

[asomers]

Summary

This can sometimes happen with broken HDDs.

MFC after: 2 weeks
Sponsored by: Axcient

Test Plan

Manually tested with a suitably broken HDD

[markj]

Seems ok, but why not handle it in g_label_taste(), or even at some higher level in GEOM (maybe g_disk_create())? I can't think of any classes that will do something sane with a provider with sectorsize 0...

[asomers]

In D37313#847880, @markj wrote:

Seems ok, but why not handle it in g_label_taste(), or even at some higher level in GEOM (maybe g_disk_create())? I can't think of any classes that will do something sane with a provider with sectorsize 0...

Ok, I'll move the part from g_label_ufs.c into g_label_taste. I don't want to move it higher up though, because g_label_taste can also be called in the resize path. And I'll leave the changes to the other label classes, because they're 0-cost.

[asomers]

• Perform the check earlier

[markj]

sys/geom/label/g_label_ufs.c

142

I'd convert the other label types to use KASSERT()s as well. Sure the runtime checks are zero-cost, but the inconsistency is confusing.

[imp]

Doh! I had hit 'looks good' too, but it didn't seem to register...